We captured and measured 1687 tadpoles o A. eugenioi in the two streams sampled. Tadpoles of A. eugenioi occurred in stream 1 during all 36 months of the study. In stream 2, tadpoles were found in 26 months and were more frequntly found in the dry season. The abundance of tadpoles in streams were heterocedastic and not normal, as it is visualized in boxplot and histogram. The analyses of the residuals of ANOVA indicated that the premisses of normality and homocedasticity were not met, so we use Kruskal-Wallis rank sum test (non-parametric ANOVA). Resutls indicated that the abundance of tadpoles difered significantly between streams.
## Class Total
## 1 1 36
## 2 2 26
## 1 2
## dry 18 15
## rainy 18 11
## Df Sum Sq Mean Sq F value Pr(>F)
## Stream 1 5805 5805 40.08 9.42e-10 ***
## Residuals 286 41426 145
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Kruskal-Wallis rank sum test
##
## data: Abundance by Stream
## Kruskal-Wallis chi-squared = 40.731, df = 1, p-value = 1.747e-10
Tadpoles of class I.2 were the most frequent being registered all months, followed by tadpoles of class II, I.1, and III. Except for tadpoles of class I.1 that was more frequent in the dry period, tadpoles of all other classes were more frequent in the rainy period.
## Class Total
## 1 I.1 23
## 2 I.2 36
## 3 II 35
## 4 III 15
## dry rainy
## I.1 14 9
## I.2 18 18
## II 17 18
## III 5 10
We captured and measured 1687 tadpoles o A. eugenioi. The most common tadpoles belonged to class I.2 and the rarer to class III. Tadpoles of all development class were more abundant in the dry period both in absolute values, by the index and the mean. The exception were tadpoles of class I.2 that were more comomn in the rainy season.
## I.1 I.2 II III
## 88 1322 251 26
## dry rainy
## I.1 60 28
## I.2 559 763
## II 165 86
## III 14 12
## dry rainy
## I.1 2.0000000 0.9333333
## I.2 18.6333333 25.4333333
## II 5.5000000 2.8666667
## III 0.4666667 0.4000000
## dry rainy
## I.1 1.6666667 0.7777778
## I.2 15.5277778 21.1944444
## II 4.5833333 2.3888889
## III 0.3888889 0.3333333
Both absolut abundances, the index, and mean abundance showed evidences of heterocedasticity and lack of normality bewteen seasons, as it is visualized in boxplot and histogram. The results of ANOVA indicated no differences in abudances between seasons. The analyses of residuals indicated that the premisses of normality and homocedasticity were not met.
## Df Sum Sq Mean Sq F value Pr(>F)
## Season 1 58 57.5 0.176 0.676
## Residuals 142 46444 327.1
## Df Sum Sq Mean Sq F value Pr(>F)
## Season 1 0.06 0.0639 0.176 0.676
## Residuals 142 51.60 0.3634
## Df Sum Sq Mean Sq F value Pr(>F)
## Season 1 14 14.38 0.176 0.676
## Residuals 142 11611 81.77
## ii) did tadpoles of different class of development tend to occur more frequently and to be more abundant in the rainy or in the dry season? ANOVA of the abundance of tadpoles of Class I.1 between seasons
The boxplot sugests the presence of outliers and heterocedasticity. The histogram indicates lack of normality as expected for counts. The test of ANOVA indicate no difference in abundances between seasons. However, plots of the residuals indicated that the premisses of normality and homocedasticity were not matched. Resutls were consistent for both absolute, the index of abundance, and mean abundance.
## Df Sum Sq Mean Sq F value Pr(>F)
## abund1$Season[abund1$Class == "I.1"] 1 28.4 28.44 2.477 0.125
## Residuals 34 390.4 11.48
##
## Kruskal-Wallis rank sum test
##
## data: abund1$Abundance[abund1$Class == "I.1"] by abund1$Season[abund1$Class == "I.1"]
## Kruskal-Wallis chi-squared = 2.9843, df = 1, p-value = 0.08408
##
## Kruskal-Wallis rank sum test
##
## data: abund.ind$Abundance[abund.ind$Class == "I.1"] by abund.ind$Season[abund.ind$Class == "I.1"]
## Kruskal-Wallis chi-squared = 2.9843, df = 1, p-value = 0.08408
## Df Sum Sq Mean Sq F value
## abund.mean$Season[abund.mean$Class == "I.1"] 1 7.11 7.111 2.477
## Residuals 34 97.61 2.871
## Pr(>F)
## abund.mean$Season[abund.mean$Class == "I.1"] 0.125
## Residuals
##
## Kruskal-Wallis rank sum test
##
## data: abund.mean$Mean[abund.mean$Class == "I.1"] by abund.mean$Season[abund.mean$Class == "I.1"]
## Kruskal-Wallis chi-squared = 2.9843, df = 1, p-value = 0.08408
The boxplot sugests the presence of heterocedasticity. The histogram indicates lack of normality as expected for counts. The test of ANOVA indicate no difference in abundances between seasons. The plots of the residuals indicated that the premisses of normality and homocedasticity were not matched. Resutls were consistent for both absolute, the index of abundance, and mean abundance.
## Df Sum Sq Mean Sq F value Pr(>F)
## abund1$Season[abund1$Class == "I.2"] 1 1156 1156.0 3.329 0.0769 .
## Residuals 34 11805 347.2
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Kruskal-Wallis rank sum test
##
## data: abund1$Abundance[abund1$Class == "I.2"] by abund1$Season[abund1$Class == "I.2"]
## Kruskal-Wallis chi-squared = 2.1217, df = 1, p-value = 0.1452
## Df Sum Sq Mean Sq F value
## abund.ind$Season[abund.ind$Class == "I.2"] 1 1.284 1.2844 3.329
## Residuals 34 13.117 0.3858
## Pr(>F)
## abund.ind$Season[abund.ind$Class == "I.2"] 0.0769 .
## Residuals
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Kruskal-Wallis rank sum test
##
## data: abund.ind$Abundance[abund.ind$Class == "I.2"] by abund.ind$Season[abund.ind$Class == "I.2"]
## Kruskal-Wallis chi-squared = 2.0304, df = 1, p-value = 0.1542
## Df Sum Sq Mean Sq F value
## abund.mean$Season[abund.mean$Class == "I.2"] 1 289 289.0 3.329
## Residuals 34 2951 86.8
## Pr(>F)
## abund.mean$Season[abund.mean$Class == "I.2"] 0.0769 .
## Residuals
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Kruskal-Wallis rank sum test
##
## data: abund.mean$Mean[abund.mean$Class == "I.2"] by abund.mean$Season[abund.mean$Class == "I.2"]
## Kruskal-Wallis chi-squared = 2.1217, df = 1, p-value = 0.1452
The boxplot sugests the presence of outliers and heterocedasticity. The histogram indicates lack of normality as expected for counts. The test of ANOVA indicate no difference in abundances between seasons. The plots of the residuals indicated that the premisses of normality and homocedasticity were not matched. Resutls were consistent for both absolute, the index of abundance, and mean abundance.
## Df Sum Sq Mean Sq F value Pr(>F)
## abund1$Season[abund1$Class == "II"] 1 173.4 173.36 2.768 0.105
## Residuals 34 2129.6 62.64
##
## Kruskal-Wallis rank sum test
##
## data: abund1$Abundance[abund1$Class == "II"] by abund1$Season[abund1$Class == "II"]
## Kruskal-Wallis chi-squared = 0.07337, df = 1, p-value = 0.7865
## Df Sum Sq Mean Sq F value Pr(>F)
## abund.ind$Season[abund.ind$Class == "II"] 1 0.1926 0.1926 2.768 0.105
## Residuals 34 2.3662 0.0696
##
## Kruskal-Wallis rank sum test
##
## data: abund.ind$Abundance[abund.ind$Class == "II"] by abund.ind$Season[abund.ind$Class == "II"]
## Kruskal-Wallis chi-squared = 0.09165, df = 1, p-value = 0.7621
## Df Sum Sq Mean Sq F value
## abund.mean$Season[abund.mean$Class == "II"] 1 43.3 43.34 2.768
## Residuals 34 532.4 15.66
## Pr(>F)
## abund.mean$Season[abund.mean$Class == "II"] 0.105
## Residuals
##
## Kruskal-Wallis rank sum test
##
## data: abund.mean$Mean[abund.mean$Class == "II"] by abund.mean$Season[abund.mean$Class == "II"]
## Kruskal-Wallis chi-squared = 0.07337, df = 1, p-value = 0.7865
The boxplot sugests the presence of outliers and heterocedasticity. The histogram indicates lack of normality as expected for counts. The test of ANOVA indicate no difference in abundances between seasons. The plots of the residuals indicated that the premisses of normality and homocedasticity were not matched. Resutls were consistent for both absolute, the index of abundance, and mean abundance.
## Df Sum Sq Mean Sq F value Pr(>F)
## abund1$Season[abund1$Class == "III"] 1 0.11 0.1111 0.074 0.787
## Residuals 34 51.11 1.5033
##
## Kruskal-Wallis rank sum test
##
## data: abund1$Abundance[abund1$Class == "III"] by abund1$Season[abund1$Class == "III"]
## Kruskal-Wallis chi-squared = 1.2843, df = 1, p-value = 0.2571
## Df Sum Sq Mean Sq F value
## abund.ind$Season[abund.ind$Class == "III"] 1 0.00012 0.0001235 0.074
## Residuals 34 0.05679 0.0016703
## Pr(>F)
## abund.ind$Season[abund.ind$Class == "III"] 0.787
## Residuals
##
## Kruskal-Wallis rank sum test
##
## data: abund.ind$Abundance[abund.ind$Class == "III"] by abund.ind$Season[abund.ind$Class == "III"]
## Kruskal-Wallis chi-squared = 1.2843, df = 1, p-value = 0.2571
## Df Sum Sq Mean Sq F value
## abund.mean$Season[abund.mean$Class == "III"] 1 0.028 0.0278 0.074
## Residuals 34 12.778 0.3758
## Pr(>F)
## abund.mean$Season[abund.mean$Class == "III"] 0.787
## Residuals
##
## Kruskal-Wallis rank sum test
##
## data: abund.mean$Mean[abund.mean$Class == "III"] by abund.mean$Season[abund.mean$Class == "III"]
## Kruskal-Wallis chi-squared = 1.2843, df = 1, p-value = 0.2571
In stream 1, tadpoles of class I.1 were the second least abundant (4.1%, n = 61). They were more frequent (30.5%, n = 11 months) and abundant (2.5%, n = 38) in the dry period. We found no statistical differences in abundances of tadpoles of class I.1 between seasons (F2.34 = 1.146, p = 0.292). Tadpoles of class I.2 occurred in all months and were the most abundant, representing 80.5% (n = 1199) of all tadpoles found in stream 1. We registered their highest abundance in the rainy season (48.5%, n = 716). Tadpoles of class I.2 were statisticaly more abundant in the rainy period (F2.34 = 4.705, p = 0.037). Tadpoles of class II the second most abundant (14.2%, n = 212) and frequent (n = 30 months). They were found more frequently during the rainy season (47.2%, n = 17), but we registered their highest abundance in the dry season (9.2%, n = 137). We did not find statistical differences in the abundances of tadpoles of class II between seasons (X2 = 0.004, p = 0.949). Tadpoles of class III (final stage of metamorphose) were the least abundant in stream 1 with only 18 individuals (1.2%) and occurred more frequnetly in the rainy season (25%, n = 9 months). We did not find significant difference in the number of tadpoles of class III between seasons (F2. = 0.569, p = 0.456).
## , , 1
##
## dry rainy
## I.1 11 7
## I.2 18 18
## II 13 17
## III 3 9
##
## , , 2
##
## dry rainy
## I.1 6 4
## I.2 13 9
## II 10 6
## III 3 1
## , , 1
##
## dry rainy
## I.1 0.30555556 0.1944444
## I.2 0.50000000 0.5000000
## II 0.36111111 0.4722222
## III 0.08333333 0.2500000
##
## , , 2
##
## dry rainy
## I.1 0.16666667 0.11111111
## I.2 0.36111111 0.25000000
## II 0.27777778 0.16666667
## III 0.08333333 0.02777778
## 1 2
## I.1 61 27
## I.2 1199 123
## II 212 39
## III 18 8
## 1 2
## I.1 0.04093960 0.018120805
## I.2 0.80469799 0.082550336
## II 0.14228188 0.026174497
## III 0.01208054 0.005369128
## 1 2
## I.1 0.30964467 0.13705584
## I.2 6.08629442 0.62436548
## II 1.07614213 0.19796954
## III 0.09137056 0.04060914
## , , 1
##
## dry rainy
## I.1 38 23
## I.2 483 716
## II 137 75
## III 7 11
##
## , , 2
##
## dry rainy
## I.1 22 5
## I.2 76 47
## II 28 11
## III 7 1
## , , 1
##
## dry rainy
## I.1 0.025503356 0.01543624
## I.2 0.324161074 0.48053691
## II 0.091946309 0.05033557
## III 0.004697987 0.00738255
##
## , , 2
##
## dry rainy
## I.1 0.014765101 0.0033557047
## I.2 0.051006711 0.0315436242
## II 0.018791946 0.0073825503
## III 0.004697987 0.0006711409
## Warning: In lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...) :
## extra argument 'p.adj' will be disregarded
## Df Sum Sq Mean Sq
## abund$Season[abund$Class == "I.1" & abund$Stream == 1] 1 6.25 6.250
## Residuals 34 185.39 5.453
## F value Pr(>F)
## abund$Season[abund$Class == "I.1" & abund$Stream == 1] 1.146 0.292
## Residuals
##
## Kruskal-Wallis rank sum test
##
## data: abund$Abundance[abund$Class == "I.1" & abund$Stream == 1] by abund$Season[abund$Class == "I.1" & abund$Stream == 1]
## Kruskal-Wallis chi-squared = 1.659, df = 1, p-value = 0.1977
## Df Sum Sq Mean Sq
## abund$Season[abund$Class == "I.2" & abund$Stream == 1] 1 1508 1508.0
## Residuals 34 10898 320.5
## F value Pr(>F)
## abund$Season[abund$Class == "I.2" & abund$Stream == 1] 4.705 0.0372 *
## Residuals
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Kruskal-Wallis rank sum test
##
## data: abund$Abundance[abund$Class == "I.2" & abund$Stream == 1] by abund$Season[abund$Class == "I.2" & abund$Stream == 1]
## Kruskal-Wallis chi-squared = 3.3708, df = 1, p-value = 0.06636
## Df Sum Sq Mean Sq
## abund$Season[abund$Class == "II" & abund$Stream == 1] 1 106.8 106.78
## Residuals 34 1562.8 45.96
## F value Pr(>F)
## abund$Season[abund$Class == "II" & abund$Stream == 1] 2.323 0.137
## Residuals
##
## Kruskal-Wallis rank sum test
##
## data: abund$Abundance[abund$Class == "II" & abund$Stream == 1] by abund$Season[abund$Class == "II" & abund$Stream == 1]
## Kruskal-Wallis chi-squared = 0.0040536, df = 1, p-value = 0.9492
## Df Sum Sq Mean Sq
## abund$Season[abund$Class == "III" & abund$Stream == 1] 1 0.444 0.4444
## Residuals 34 26.556 0.7810
## F value Pr(>F)
## abund$Season[abund$Class == "III" & abund$Stream == 1] 0.569 0.456
## Residuals
##
## Kruskal-Wallis rank sum test
##
## data: abund$Abundance[abund$Class == "III" & abund$Stream == 1] by abund$Season[abund$Class == "III" & abund$Stream == 1]
## Kruskal-Wallis chi-squared = 3.1458, df = 1, p-value = 0.07612
In stream 2, tadpoles of class I.1 were the second least abundant (13.7%, n = 27). They were more frequent (16.6%, n = 6 months) and abundant (11.2%, n = 22) in the dry period. We found no statistical differences in abundances of tadpoles of class I.1 between seasons (X2 = 0.818, p = 0.365). Tadpoles of class I.2 occurred were the most abundant, representing 62.4% (n = 123) of all tadpoles found in stream 2. They were more frequent (36.1%, n = 13) and abundant in the dry season (38.6%, n = 76). Tadpoles of class I.2 did not differ statisticaly between seasons (F2.34 = 1.183, R2 = 0.005, p = 0.284). Tadpoles of class II the second most abundant (19.8%, n = 39). They were more abundant (14.2%, n = 28) and found more frequently during the dry season (27.7%, n = 10). We did not find statistical differences in the abundances of tadpoles of class II between seasons (X2 = 1.837, p = 0.175). Tadpoles of class III (final stage of metamorphose) were the least abundant in stream 2 with only 8 individuals (4.1%) and occurred more frequnetly in the dry season (8.3%, n = 3 months) when they were also more abundant (3.5%, n = 7). We did not find significant difference in the number of tadpoles of class III between seasons (X2 = 1.1524, p = 0.283).
## , , 1
##
## dry rainy
## I.1 11 7
## I.2 18 18
## II 13 17
## III 3 9
##
## , , 2
##
## dry rainy
## I.1 6 4
## I.2 13 9
## II 10 6
## III 3 1
## , , 1
##
## dry rainy
## I.1 0.30555556 0.1944444
## I.2 0.50000000 0.5000000
## II 0.36111111 0.4722222
## III 0.08333333 0.2500000
##
## , , 2
##
## dry rainy
## I.1 0.16666667 0.11111111
## I.2 0.36111111 0.25000000
## II 0.27777778 0.16666667
## III 0.08333333 0.02777778
## 1 2
## I.1 61 27
## I.2 1199 123
## II 212 39
## III 18 8
## 1 2
## I.1 0.04093960 0.018120805
## I.2 0.80469799 0.082550336
## II 0.14228188 0.026174497
## III 0.01208054 0.005369128
## 1 2
## I.1 0.30964467 0.13705584
## I.2 6.08629442 0.62436548
## II 1.07614213 0.19796954
## III 0.09137056 0.04060914
## , , 1
##
## dry rainy
## I.1 38 23
## I.2 483 716
## II 137 75
## III 7 11
##
## , , 2
##
## dry rainy
## I.1 22 5
## I.2 76 47
## II 28 11
## III 7 1
## , , 1
##
## dry rainy
## I.1 0.19289340 0.11675127
## I.2 2.45177665 3.63451777
## II 0.69543147 0.38071066
## III 0.03553299 0.05583756
##
## , , 2
##
## dry rainy
## I.1 0.11167513 0.025380711
## I.2 0.38578680 0.238578680
## II 0.14213198 0.055837563
## III 0.03553299 0.005076142
## Warning: In lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...) :
## extra argument 'p.adj' will be disregarded
## Df Sum Sq Mean Sq
## abund$Season[abund$Class == "I.1" & abund$Stream == 2] 1 8.03 8.028
## Residuals 34 134.72 3.962
## F value Pr(>F)
## abund$Season[abund$Class == "I.1" & abund$Stream == 2] 2.026 0.164
## Residuals
##
## Kruskal-Wallis rank sum test
##
## data: abund$Abundance[abund$Class == "I.1" & abund$Stream == 2] by abund$Season[abund$Class == "I.1" & abund$Stream == 2]
## Kruskal-Wallis chi-squared = 0.81878, df = 1, p-value = 0.3655
## Df Sum Sq Mean Sq
## abund$Season[abund$Class == "I.2" & abund$Stream == 2] 1 23.4 23.36
## Residuals 34 671.4 19.75
## F value Pr(>F)
## abund$Season[abund$Class == "I.2" & abund$Stream == 2] 1.183 0.284
## Residuals
##
## Call:
## lm(formula = abund$Abundance[abund$Class == "I.2" & abund$Stream ==
## 2] ~ abund$Season[abund$Class == "I.2" & abund$Stream ==
## 2])
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.222 -2.611 -1.417 1.931 10.778
##
## Coefficients:
## Estimate
## (Intercept) 4.222
## abund$Season[abund$Class == "I.2" & abund$Stream == 2]rainy -1.611
## Std. Error
## (Intercept) 1.047
## abund$Season[abund$Class == "I.2" & abund$Stream == 2]rainy 1.481
## t value
## (Intercept) 4.031
## abund$Season[abund$Class == "I.2" & abund$Stream == 2]rainy -1.088
## Pr(>|t|)
## (Intercept) 0.000296 ***
## abund$Season[abund$Class == "I.2" & abund$Stream == 2]rainy 0.284389
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.444 on 34 degrees of freedom
## Multiple R-squared: 0.03363, Adjusted R-squared: 0.005202
## F-statistic: 1.183 on 1 and 34 DF, p-value: 0.2844
##
## Kruskal-Wallis rank sum test
##
## data: abund$Abundance[abund$Class == "I.2" & abund$Stream == 2] by abund$Season[abund$Class == "I.2" & abund$Stream == 2]
## Kruskal-Wallis chi-squared = 1.581, df = 1, p-value = 0.2086
## Df Sum Sq Mean Sq
## abund$Season[abund$Class == "II" & abund$Stream == 2] 1 8.03 8.028
## Residuals 34 168.72 4.962
## F value Pr(>F)
## abund$Season[abund$Class == "II" & abund$Stream == 2] 1.618 0.212
## Residuals
##
## Kruskal-Wallis rank sum test
##
## data: abund$Abundance[abund$Class == "II" & abund$Stream == 2] by abund$Season[abund$Class == "II" & abund$Stream == 2]
## Kruskal-Wallis chi-squared = 1.8378, df = 1, p-value = 0.1752
## Df Sum Sq Mean Sq
## abund$Season[abund$Class == "III" & abund$Stream == 2] 1 1.00 1.0000
## Residuals 34 25.22 0.7418
## F value Pr(>F)
## abund$Season[abund$Class == "III" & abund$Stream == 2] 1.348 0.254
## Residuals
##
## Kruskal-Wallis rank sum test
##
## data: abund$Abundance[abund$Class == "III" & abund$Stream == 2] by abund$Season[abund$Class == "III" & abund$Stream == 2]
## Kruskal-Wallis chi-squared = 1.1524, df = 1, p-value = 0.2831
Tadpoles varied in body length and relative developmental stages along the study period in both streams (Figures 2A and 2B). Mean body length of tadpoles A. eugenioi from stream 1 were usually larger (Table 2). Class I.1 tadpoles of stream 1 were significantly larger in their body size in the first and second years (Mann-Whitney; U year 1 = 11686.5; pyear 1 = 0.01; Uyear2 = 18043; p year2 < 0.001), despites of the two maximum body sizes had occurred mainly in tadpoles of stream 2 (Table 2; Figures 2A and 2B). At least in stream 1, we monthly found tadpoles smaller than 7.0 mm (Figure 3). In class II tadpoles, the largest minimum body sizes of tadpoles were found in stream 1, while the largest maximum body sizes were encountered mainly in stream 2 (Table 2; Figures 2A and 2B). We did not find significant differences in body length of class II tadpoles between the two streams, although their body lengths were generally larger in stream 1 (Table 2; Figures 2A and 2B). Tadpoles of class III were usually larger at stream 1, except in the third year (Table 2; Figures 2A and 2B). Larger We registered minimum body sizes at class III in stream 1, and larger maximum body sizes mainly in stream 2 (Table 2; Figures 3 and 4). Statistical differences in tadpoles of class III could not be tested due to the small sample size.
## 1 2
## I.1 4.575410 3.929630
## I.2 8.871005 8.346281
## II 12.044498 11.430952
## III 14.633333 13.957143
## 1 2
## I.1 0.4299828 0.7378841
## I.2 2.2962050 2.0550361
## II 2.0943332 2.4707546
## III 1.0748461 2.3042404
## 1 2
## I.1 5.0 5.0
## I.2 15.9 16.6
## II 16.5 16.8
## III 17.5 17.6
## 1 2
## I.1 2.7 2.6
## I.2 5.1 5.1
## II 6.4 6.9
## III 13.0 11.9
## Df Sum Sq Mean Sq F value
## bd.length$Stream[bd.length$Class == "I.1"] 1 7.805 7.805 26.58
## Residuals 86 25.249 0.294
## Pr(>F)
## bd.length$Stream[bd.length$Class == "I.1"] 1.59e-06 ***
## Residuals
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 10 observations deleted due to missingness
##
## Call:
## lm(formula = bd.length$`Body length`[bd.length$Class == "I.1"] ~
## bd.length$Stream[bd.length$Class == "I.1"])
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.87541 -0.37541 0.07459 0.32459 1.07037
##
## Coefficients:
## Estimate Std. Error t value
## (Intercept) 4.57541 0.06938 65.951
## bd.length$Stream[bd.length$Class == "I.1"]2 -0.64578 0.12525 -5.156
## Pr(>|t|)
## (Intercept) < 2e-16 ***
## bd.length$Stream[bd.length$Class == "I.1"]2 1.59e-06 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.5418 on 86 degrees of freedom
## (10 observations deleted due to missingness)
## Multiple R-squared: 0.2361, Adjusted R-squared: 0.2272
## F-statistic: 26.58 on 1 and 86 DF, p-value: 1.592e-06
## Warning: In lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...) :
## extra argument 'main' will be disregarded
##
## Kruskal-Wallis rank sum test
##
## data: bd.length$`Body length`[bd.length$Class == "I.1"] by bd.length$Stream[bd.length$Class == "I.1"]
## Kruskal-Wallis chi-squared = 14.725, df = 1, p-value = 0.0001244
## Df Sum Sq Mean Sq F value
## bd.length$Stream[bd.length$Class == "I.2"] 1 30 30.227 5.84
## Residuals 1303 6744 5.176
## Pr(>F)
## bd.length$Stream[bd.length$Class == "I.2"] 0.0158 *
## Residuals
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 10 observations deleted due to missingness
##
## Call:
## lm(formula = bd.length$`Body length`[bd.length$Class == "I.2"] ~
## bd.length$Stream[bd.length$Class == "I.2"])
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.771 -1.871 -0.171 1.654 8.254
##
## Coefficients:
## Estimate Std. Error t value
## (Intercept) 8.87101 0.06612 134.170
## bd.length$Stream[bd.length$Class == "I.2"]2 -0.52472 0.21714 -2.417
## Pr(>|t|)
## (Intercept) <2e-16 ***
## bd.length$Stream[bd.length$Class == "I.2"]2 0.0158 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.275 on 1303 degrees of freedom
## (10 observations deleted due to missingness)
## Multiple R-squared: 0.004462, Adjusted R-squared: 0.003698
## F-statistic: 5.84 on 1 and 1303 DF, p-value: 0.0158
## Warning: In lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...) :
## extra argument 'main' will be disregarded
##
## Kruskal-Wallis rank sum test
##
## data: bd.length$`Body length`[bd.length$Class == "I.2"] by bd.length$Stream[bd.length$Class == "I.2"]
## Kruskal-Wallis chi-squared = 5.4971, df = 1, p-value = 0.01905
## Df Sum Sq Mean Sq F value
## bd.length$Stream[bd.length$Class == "II"] 1 13.2 13.165 2.82
## Residuals 249 1162.6 4.669
## Pr(>F)
## bd.length$Stream[bd.length$Class == "II"] 0.0944 .
## Residuals
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 10 observations deleted due to missingness
##
## Call:
## lm(formula = bd.length$`Body length`[bd.length$Class == "II"] ~
## bd.length$Stream[bd.length$Class == "II"])
##
## Residuals:
## Min 1Q Median 3Q Max
## -5.6445 -1.3945 0.2555 1.5055 5.3690
##
## Coefficients:
## Estimate Std. Error t value
## (Intercept) 12.0445 0.1495 80.583
## bd.length$Stream[bd.length$Class == "II"]2 -0.6135 0.3654 -1.679
## Pr(>|t|)
## (Intercept) <2e-16 ***
## bd.length$Stream[bd.length$Class == "II"]2 0.0944 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.161 on 249 degrees of freedom
## (10 observations deleted due to missingness)
## Multiple R-squared: 0.0112, Adjusted R-squared: 0.007225
## F-statistic: 2.82 on 1 and 249 DF, p-value: 0.09438
## Warning: In lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...) :
## extra argument 'main' will be disregarded
##
## Kruskal-Wallis rank sum test
##
## data: bd.length$`Body length`[bd.length$Class == "II"] by bd.length$Stream[bd.length$Class == "II"]
## Kruskal-Wallis chi-squared = 3.7987, df = 1, p-value = 0.05129
## Df Sum Sq Mean Sq F value
## bd.length$Stream[bd.length$Class == "III"] 1 2.3 2.304 1.029
## Residuals 23 51.5 2.239
## Pr(>F)
## bd.length$Stream[bd.length$Class == "III"] 0.321
## Residuals
## 10 observations deleted due to missingness
##
## Call:
## lm(formula = bd.length$`Body length`[bd.length$Class == "III"] ~
## bd.length$Stream[bd.length$Class == "III"])
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.0571 -1.2333 -0.0333 0.4667 3.6429
##
## Coefficients:
## Estimate Std. Error t value
## (Intercept) 14.6333 0.3527 41.491
## bd.length$Stream[bd.length$Class == "III"]2 -0.6762 0.6665 -1.015
## Pr(>|t|)
## (Intercept) <2e-16 ***
## bd.length$Stream[bd.length$Class == "III"]2 0.321
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.496 on 23 degrees of freedom
## (10 observations deleted due to missingness)
## Multiple R-squared: 0.04283, Adjusted R-squared: 0.001217
## F-statistic: 1.029 on 1 and 23 DF, p-value: 0.3209
## Warning: In lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...) :
## extra argument 'main' will be disregarded
##
## Kruskal-Wallis rank sum test
##
## data: bd.length$`Body length`[bd.length$Class == "III"] by bd.length$Stream[bd.length$Class == "III"]
## Kruskal-Wallis chi-squared = 1.6179, df = 1, p-value = 0.2034